OLED could provide visual boost next Quest Pros need
The Apple Vision Pro might be the talk of the town in the high-end VR space right now, but not only will it have to contend with rivals like the Samsung XR headset that are expected to launch later this year, new reports claim the Meta Quest Pro 2 will launch in early 2025 to compete with it too.
The original Meta Quest Pro was something of a disappointment. At the time it seemed like a decent option for people looking for a high-end standalone VR headset – especially compared to rivals like the HTC Vive XR Elite. But since the launch of the Meta Quest 3 and Vision Pro – the former of which is not only cheaper but actually has some better specs in the mixed-reality department – it’s fallen by the wayside.
Meta is clearly hoping to make its next Quest Pro device a standout VR gadget. A Korea Economic Daily report (translated from Korean) cites unnamed industry sources who said Meta CEO Mark Zuckerberg is set to meet with the CEO of LG Electronics to discuss a partnership for its next Pro devices.
While we hope these high-end screens will make their way to the more budget-friendly Quest line it’s more reasonable to assume that a pricier Quest Pro line would be upgraded to LG OLEDs first.
LG makes fantastic OLED TVs like the LG C3 (Image credit: LG)
Yes, we know the original Oculus Quest got there first, but since then Meta has relied on LCD panels with better brightness and resolution because the original OLED Quest couldn’t benefit from this display tech’s full advantages. The pixels took too long to turn on and off so you could never experience true blacks, despite the ability to achieve true blacks being the main reason to use an OLED.
Thankfully, the recent Korea Economic Daily report said we might not be waiting too long. The Meta Quest Pro 2 is apparently being prepared for an early 2025 launch, and while this is a slight departure from Meta’s usual October release strategy it makes some sense.
However, as with all rumors, we must remember to take these reports with a pinch of salt. Until Meta or LG make an official announcement there's no guarantee they’re working together on the next Quest Pro or any kind of headset – nor a guarantee of when it’ll launch and what specs it might have.
As soon as we do hear anything more concrete, or we spy any interesting leaks and rumors, we’ll be sure to keep you informed.
Google is gearing up to relaunch its image creation tool that’s part of the newly-rebranded generative artificial intelligence (AI) bot, Gemini, in the next few weeks. The generative AI image creation tool is in theory capable of generating almost anything you can dream up and put into words as a prompt, but “almost” is the key word here.
Google has pumped the brakes on Gemini’s image generation after Gemini was observed creating historical depictions and other questionable images that were considered inaccurate or offensive. However, it looks like Gemini could return to image generation soon, as Google DeepMind CEO Demis Hassabis announced that Gemini will be rebooted in the coming week after taking time to address these issues.
Image generation came to Gemini earlier in February, and users were keen to test its abelites. Some people attempted to generate images depicting a certain historical period that appeared to greatly deviate from accepted historical fact. Some of these users took to social media to share their results and direct criticism at Google.
The images caught many people’s attention and sparked many conversations, and Google has recognized the images as a symptom of a problem within Gemini. The tech giant then chose to take the feature offline and fix whatever was causing the model to dream up such strange and controversial pictures.
Hassabis confirmed that Gemini was not working as intended, and that it would take some weeks to amend it, and bring it back online while speaking at a panel taking place at the Mobile World Congress (MWC) event in Barcelona.
(Image credit: Shutterstock)
If at first, your generative AI bot doesn't succeed…
Google’s first attempt at a generative AI chatbot was Bard, which saw a lukewarm reception and didn’t win users over from the more popular ChatGPT in the way Google had hoped, after which it changed course and debuted its revamped and rebranded family of generative models, Gemini. Like ChatGPT, Google is now offering a premium-tier for Gemini, which offers advanced features for a subscription.
The examples of Gemini's misadventures have also reignited discussions about AI ethics generally, and Google’s AI ethics specifically, and around issues like the accuracy of generated AI output and AI hallucinations. Companies like Microsoft and Google are pushing ahead to win the AI assistant arms race, but while racing ahead, they’re in danger of releasing products with flaws that could undermine their hard work.
AI-generated content is becoming increasingly popular and, especially due to their size and resources, these companies can (and really, should) be held to a high standard of accuracy. High profile fails like the one Gemini experienced aren’t just embarrassing for Google – it could damage the product’s perception in the eyes of consumers. There’s a reason Google rebranded Bard after its much-mocked debut.
There’s no doubt that AI is incredibly exciting, but Google and its peers should be mindful that rushing out half-baked products just to get ahead of the competition could spectacularly backfire.
Nvidia’s free tier of GeForce Now, its cloud gaming service, will soon run up to two minutes of ads before you play, according to Nvidia spokesperson Stephanie Ngo.
GeForce Now is a service offered by Nvidia that allows you to connect to digital PC game stores and stream games you already own across a multitude of different devices – including Macs, Windows laptops, iPhones and iPads, Android phones, and more.
It offers three membership tiers, with the free membership offering a queue system with an hour-long gaming session length that will then bring you back to the start of the queue once your time is up. It’s in this waiting time that the ads will be shown, so while it could be a little annoying, your actual gameplay time won’t be interrupted.
The ads will help pay for the free tier service and keep it free, with Ngo adding that the change is also expected to reduce wait times for free users in the long run – though it’s not entirely clear at this point how that’s going to work. Perhaps Nvidia is expecting the arrival of ads to push users to pay for the premium tiers or simply drive some users away from the platform entirely – either would, in theory, help reduce queues for the free tier. GeForce Now users should expect an email on 27 Feb to let them know about the changes.
Major inconvenience or just … meh?
I’m not a user of Nvidia’s game-streaming service myself, but I reached out to GeForce Now Members within the TechRadar team and learned that wait times currently fluctuate between five to fifteen minutes – and scrolling through the GeForce Now subreddit proves that wait times can go on even longer.
Most people who use the free tier of GeForce Now go in aware that they will be spending a not-insignificant amount of time in a queue, so in reality, two minutes of ads when you know you’re likely going to be waiting for longer anyway isn’t much of an inconvenience – it might even help kill some time. Many users are likely to simply do something else while queuing for their free hour timeslot anyway, so why shouldn’t Nvidia get some extra ad revenue from it?
That being said, it is a gloomy example of the inescapable modern torture of being advertised at non-stop. Almost every facet of the internet is packed with ads at this point (this article included – sorry about that, but we’ve got to eat!) and while a lot of platforms offer ad-free paid tiers, it seems like that isn’t enough anymore.
Amazon Prime has received a lot of (well-deserved) flak for slapping ads onto paid memberships, and Netflix’s ad-supported free tier wasn’t very well-received either. While Nvidia’s latest move seems fairly innocuous right now, who’s to say the ‘up to two minutes’ won’t extend further in the future, until you’re sat watching a full ten minutes of commercials to play an hour-long session of your current favorite game? Do you just give in and buy a paid membership? I just might, personally – but I wouldn’t be happy about it.
Windows 11 is now adding support for Wi-Fi 7, those who want to use the much-improved wireless standard will doubtless be pleased to learn – but it’s only in testing currently.
As you might guess, it’ll be some time before official Wi-Fi 7 support comes through to the release version of Windows 11, as it’ll need to progress through testing channels first.
Right now, it’s only in the Canary (earliest) test channel with build 26063, a preview release that flew under our radar somewhat, but an important one in this respect. However, it’s also been added for Dev channel testers, Microsoft informed us in the usual blog post on build 26063 (as flagged up by XDA Developers).
(Image credit: Microsoft)
As the software giant also pointed out, Wi-Fi 7 (aka 802.11be) is in the order of 4x faster than Wi-Fi 6 and more like 6x quicker than Wi-Fi 5.
In fairness to Microsoft, while it appears to be pretty late to the wireless party, and Wi-Fi 7 may have officially kicked off (at least in some countries, the US, Australia, and UK included), it's still early days for the standard.
The standard may be effectively set in stone now, but that doesn’t mean there won’t be tweaks going forward. There will inevitably be firmware updates for existing Wi-Fi 7 routers to fix or modify things going forward as needed, although all the big cogs in terms of features are now in place.
Windows 11 is one of the final pieces of the puzzle to be added for Wi-Fi 7 support, then, for laptops which sport Wi-Fi 7. And of course as mentioned, you’ll need a Wi-Fi 7 router to benefit from faster wireless speeds. (Those devices are expensive right now, too, it should be noted – though that’s generally true of any cutting-edge tech).
With Wi-Fi 7 we’re getting performance which makes wireless online gaming a reality in terms of it being close to wired (Ethernet) performance, and certainly much better than other fudges for PCs that aren’t plugged directly into the router (such as powerline adapters, which can be notoriously flaky in some scenarios).
As part of a recent Android feature drop, the Fitbit app will now show data from third-party sources to provide users with a “more complete picture of [their] health”.
Google stated in a recent announcement it’s effectively expanding Health Connect’s reach, allowing it to grab stats “from your favorite wearables and apps”. These sources include AllTrials, the Oura Ring, plus nutritional data from MyFitnessPal. Over in the Today tab on the Fitbit app, you will see a new section called Records where you will see all the Health Connect info listed out in detail.
Looking at the demo video, there are entries for calories burned in a day, distance traveled, floors climbed, and body measurements among other things. Tapping an entry will take you to a stat readout. For example, going to “Steps” will show you how many steps you’ve taken in a day, week, month, and year with a daily average number on the side.
Data coming from a third-party source will have the service’s logo right next to it. The aforementioned Steps section has a symbol of an Oura Ring next to it while Elevation Gained has the AllTrials icon alongside it. It’s important to mention that there may be a discrepancy in the information shown. 9To5Google explains in its coverage “data in Health Connect may not match the metrics you see on your Fitbit devices.”
Android feature drop
The update is currently rolling out alongside the eight other features. Just to briefly go over them, you have WearOS smartwatches receiving public transit directions via Google Maps plus support for Google Wallet passes. Google Messages will soon host the company’s Gemini chatbot so you can have direct conversations with the AI. And the Android home screen will gain an output switcher for Spotify giving subscribers the ability to change where their media is playing. You’ll be able to seamlessly hop between a smartphone, a pair of headphones, or a smart TV.
We reached out to Google asking if the only third-party sources Health Connect has access to are All Trials, an Oura Ring, and MyFitnessPal or if there are more. This story will be updated at a later time.
Pretty much anything we can do with AI today might have seemed like magic just a year ago, but MindStudio's platform for creating custom AI apps in a matter of minutes feels like a new level of alchemy.
The six-month-old free platform, which you can find right now under youai.ai, is a visual studio for building AI workflows, assistants, and AI chatbots. In its short lifespan it's already been used, according to CEO Dimitry Shapiro, to build more than 18,000 apps.
Yes, he called them “apps”, and if you're struggling to understand how or why anyone might want to build AI applications, just look at OpenAI's relatively new GPT apps (aka GPTs). These let you lock the powerful GPT-3.5 into topic-based thinking that you can package up, share, and sell. Shapiro, however, noted the limits of OpenAI's approach.
He likened GPTs to “bookmarking a prompt” within the GPT sphere. MindStudio, on the other hand, is generative model-agnostic. The system lets you use multiple models within one app.
If adding more model options sounds complicated, I can assure you it's not. MindStudio is the AI development platform for non-developers.
Watch and learn
Choose your template. (Image credit: MindStudio)
To get you started, the company provides an easy-to-follow 18-minute video tutorial. The system also helps by offering a healthy collection of templates (many of them business-focused), or you can choose a blank template. I followed the guide to recreate the demo AI app (a blog post generator), and my only criticism is that the video is slightly out of date, with some interface elements having been moved or renamed. There are some prompts to note the changes, but the video could still do with a refresh.
Still, I had no trouble creating that first AI blog generator. The key here is that you can get a lot of the work done through a visual interface that lets you add blocks along a workflow and then click on them to customize, add details, and choose which AI model you want to use (the list includes GPT- 3.5 turbo, PaLM 2, Llama 2, and Gemini Pro). While you don't necessarily have to use a particular model for each task in your app, it might be that, for example, you should be using GPT-3.5 for fast chatbots or that PaLM would be better for math; however, MindStudio cannot, at least yet, recommend which model to use and when.
Image 1 of 2
Connect the boxes (Image credit: MindStudio)
Image 2 of 2
And then edit their contents (Image credit: MindStudio)
The act of adding training data is also simple. I was able to find web pages of information, download the HTML, and upload it to MindStudio (you can upload up to 150 files on a single app). MindStudio uses the information to inform the AI, but will not be cutting and pasting information from any of those pages into your app responses.
Most of MindStudio's clients are in business, and it does hide some more powerful features (embedding on third-party websites) and models (like GPT 4 Turbo) behind a paywall, but anyone can try their hand at building and sharing AI apps (you get a URL for sharing).
Confident in my newly acquired, if limited, knowledge, I set about building an AI app revolving around mobile photography advice. Granted, I used the framework I'd just learned in the AI blog post generator tutorial, but it still went far better than I expected.
One of the nice things about MindStudio is that it allows for as much or as little coding as you're prepared to do. In my case, I had to reference exactly one variable that the model would use to pull the right response.
Options include setting your model’s ‘temperature’ (Image credit: MindStudio)
There are a lot of smart and dead-simple controls that can even teach you something about how models work. MindStudio lets you set, for instance, the 'Temperature' of your model to control the randomness of its responses. The higher the 'temp', the more unique and creative each response. If you like your model verbose, you can drag another slider to set a response size of up to 3,000 characters.
The free service includes unlimited consumer usage and messages, some basic metrics, and the ability to share your AI via a link (as I've done here). Pro users can pay $ 23 a month for the more powerful models like GPT-4, less MindStudio branding, and, among other things, site embedding. The $ 99 a-month tier includes all you get with Pro, but adds the ability to charge for access to your AI app, better analytics, API access, full chat transcripts, and enterprise support.
Image 1 of 2
Look, may, I made an AI app. (Image credit: MindStudio)
Image 2 of 2
It’s smarter than I am. (Image credit: MindStudio)
I can imagine small and medium-sized businesses using MindStudio to build customer engagement and content capture on their sites, and even as a tool for guiding users through their services.
Even at the free level, though, I was surprised at the level of customization MindStorm offers. I could add my own custom icons and art, and even build a landing page.
I wouldn't call my little AI app anything special, but the fact that I could take the germ of an idea and turn it into a bespoke chatbot in 10 minutes is surprising even to me. That I get to choose the right model for each job within an AI app is even better; and that this level of fun and utility is free is the icing on the cake.
A few weeks ago Google launched a new image generation tool for Gemini (the suite of AI tools formerly known as Bard and Duet) which allowed users to generate all sorts of images from simple text prompts. Unfortunately, Google’s AI tool repeatedly missed the mark and generated inaccurate and even offensive images that led a lot of us to wonder – how did the bot get things so wrong? Well, the company has finally released a statement explaining what went wrong, and how it plans to fix Gemini.
The official blog post addressing the issue states that when designing the text-to-image feature for Gemini, the team behind Gemini wanted to “ensure it doesn’t fall into some of the traps we’ve seen in the past with image generation technology — such as creating violent or sexually explicit images, or depictions of real people.” The post further explains that users probably don’t want to keep seeing people of just one ethnicity or other prominent characteristic.
So, to offer a pretty basic explanation for what’s been going on: Gemini has been throwing up images of people of color when prompted to generate images of white historical figures, giving users ‘diverse Nazis’, or simply ignoring the part of your prompt where you’ve specified exactly what you’re looking for. While Gemini’s image capabilities are currently on hold, when you could access the feature you’d specify exactly who you’re trying to generate – Google uses the example “a white veterinarian with a dog” – and Gemini would seemingly ignore the first half of that prompt and generate veterinarians of all races except the one you asked for.
Google went on to explain that this was the outcome of two crucial failings – firstly, Gemini was showing a range of different people without considering a range not to show. Alongside that, in trying to make a more conscious, less biased generative AI, Google admits the “model became way more cautious than we intended and refused to answer certain prompts entirely – wrongly interpreting some very anodyne prompts as sensitive.”
So, what's next?
At the time of writing, the ability to generate images of people on Gemini has been paused while the Gemini team works to fix the inaccuracies and carry out further testing. The blog post notes that AI ‘hallucinations’ are nothing new when it comes to complex deep learning models – even Bard and ChatGPT had some questionable tantrums as the creators of those bots worked out the kinks.
The post ends with a promise from Google to keep working on Gemini’s AI-powered people generation until everything is sorted, with the note that while the team can’t promise it won’t ever generate “embarrassing, inaccurate or offensive results”, action is being taken to make sure it happens as little as possible.
All in all, this whole episode puts into perspective that AI is only as smart as we make it. Our editor-in-chief Lance Ulanoff succinctly noted that “When an AI doesn't know history, you can't blame the AI.” With how quickly artificial intelligence has swooped in and crammed itself into various facets of our daily lives – whether we want it or not – it’s easy to forget that the public proliferation of AI started just 18 months ago. As impressive as the tools currently available to us are, we’re ultimately still in the early days of artificial intelligence.
We can’t rain on Google Gemini’s parade just because the mistakes were more visually striking than say, ChatGPT’s recent gibberish-filled meltdown. Google’s temporary pause and reworking will ultimately lead to a better product, and sooner or later we’ll see the tool as it was meant to be.
The Oculus Quest 2 has been the most popular VR headset in the world for the past couple of years – dominating sales and usage charts with its blend of solid performance, amazing software library and, most importantly, affordability.
Now its successor – the Meta Quest 3 – is following in its footsteps.
Just four months after launch it’s the third most popular headset used on Steam (and will likely be the second most popular in the next Steam Hardware Survey). What’s more, while we estimate the Quest 3’s not selling quite as well as the Quest 2 was at the four-month mark, it still looks to be a hit (plus, lower sales figures are expected considering it’s almost double the launch price of the Quest 2).
Despite its higher cost, $ 499.99 / £479.99 / AU$ 799.99 is still relatively affordable in the VR space, and its early success continues the ongoing trend in VR that accessibility is the make or break factor in a VR gadget’s popularity.
The cheap Oculus Quest 2 made VR mainstream (Image credit: Facebook)
There’s something to be said for high-end hardware such as the Apple Vision Pro bringing the wow factor back to VR (how can you not be impressed by its crisp OLED displays and inventive eye-and-hand-tracking system), but I’ll admit I was worried that its launch – and announcement of other high-end, and high-priced, headsets – would see VR return to its early, less affordable days.
Now I’m more confident than ever that we’ll see Apple’s rumored cheaper Vision Pro follow-up and other budget-friendly hardware sooner rather than later.
Rising up the charts
According to the Steam Hardware Survey, which tracks the popularity of hardware for participating Steam users, 14.05% of all Steam VR players used a Quest 3 last month. That’s a 4.78% rise in its popularity over the previous month’s results and means it’s within spitting distance of the number two spot, which is currently held by the Valve Index – 15% of users prefer it over other VR headsets, even three-and-a-half years after its launch.
It has a ways to go before it reaches the top spot, however, with the Oculus Quest 2 preferred by 40.64% of Steam VR players. The Quest 3’s predecessor has held this top spot for a couple of years now, and it’s unlikely to lose to the Quest 3 or another headset for a while. Even though the Quest 3 is doing well for itself, it’s not selling quite as fast as the Quest 2.
(Image credit: Future)
Using Steam Hardware Survey data for January 2024 (four months after its launch) and data from January 2021 (four months after the Quest 2’s launch) – as well as average Steam player counts for these months based on SteamDB data – it appears that the Quest 3 has sold about 87% as many units as the Quest 2 did at the same point in its life.
Considering the Quest 3 is priced at $ 499.99 / £479.99 / AU$ 799.99, a fair bit more than the $ 299 / £299 / AU$ 479 the Quest 2 cost at launch, to even come close to matching the sales speed of its predecessor is impressive. And the Quest 2 did sell very well out of the gate.
We don’t have exact Quest 2 sales data from its early days – Meta only highlights when the device passes certain major milestones – but we do know that after five months, its total sales were higher than the total sales of all other Oculus VR headsets combined, some of which had been out for over five years. Meta’s gone on to sell roughly 20 million Quest 2s, according to a March 2023 leak. That's about as fast as the Xbox Series X is believed to have sold, which launched around the same time.
This 87% of Quest 2 sales figure can be taken with a pinch of salt – you can find out how I got to this number at the bottom of this piece; it required pulling data from a few sources and making some reasonable assumptions – but that number and the Quest 2 and 3’s popularity on Steam shows that affordability is still the most powerful driving force in the VR space. So, I hope other headset makers are paying attention.
The Apple Vision had me a little concerned (Image credit: Future)
The early popularity makes sense. Whatever Mark Zuckerberg says about the superiority of the Quest 3, the Apple Vision Pro is the best of the best VR headsets from a technical perspective. There’s some debate on the comfort and immersive software side of things, but eye-tracking, ridiculously crisp OLED displays, and a beautiful design do make up for that.
Unfortunately, thanks to these high-end specs and some ridiculous design choices – like the outer OLED display for EyeSight (which lets an onlooker see the wearer’s eyes while they're wearing the device) – the headset is pretty pricey coming in at $ 3,499 for the 256GB model (it’s not yet available outside the US).
Seeing this, and the instant renewed attention Apple has drawn to the VR space – with high-end rivals like the Samsung XR headset now on the way – I’ll admit I was a little concerned we might see a return to VR’s early, less accessible days. In those days, you’d spend around $ 1,000 / £1,000 / AU$ 1,500 on a headset and the same again (or more) on a VR-ready PC.
The Valve Index is impressive, but it’s damn expensive (Image credit: Future)
Apple has a way of driving the tech conversation and development in the direction it chooses. Be it turning more niche tech into a mainstream affair like it did for smartwatches with the Apple Watch or renaming well-established terms by sheer force of will (VR computing and 3D video are now exclusively called spatial computing and spatial video after Apple started using those phrases).
While, yes, there’s something to be said for the wow factor of top-of-the-line tech, I hoped we wouldn’t be swamped with the stuff while more budget-friendly options get forgotten about because this is the way Apple has moved the industry with its Vision Pro.
The numbers in the Steam Hardware Survey have assuaged those fears. It shows that meaningful budget hardware – like the Quest 2 and 3, which, despite being newer, have less impressive displays and specs than many older, pricier models – is still too popular to be going anywhere anytime soon.
If anything, I’m more confident than ever that Apple, Samsung, and the like need to get their own affordable VR headsets out the door soon. Especially the non-Apple companies that can’t rely on a legion of rabid fans ready to eat up everything they release.
If they don’t launch budget-friendly – but still worthwhile – VR headsets, then Meta could once again be left as the only real contender in this sector of VR. Sure, I like the Meta headsets I’ve used, but nothing helps spur on better tech and/or prices than proper competition. And this is something Meta is proving it doesn’t really have right now.
(Image credit: Meta)
Where did my data come from?
It’s important to know where data has come from and what assumptions have been made by people handling that data, but, equally, not everyone finds this interesting, and it can get quite long and distracting. So, I’ve put this section at the bottom for those interested in seeing my work on the 87% sales figure comparison between the Oculus Quest 2 and Meta Quest 3 four months after their respective launches.
As I mentioned above, most of the data for this piece has been gathered from the Steam Hardware Survey. I had to rely on the Internet Archive’s Wayback Machine to see some historical Steam Hardware Survey data because the results page only shows the most recent month’s figures.
When looking at the relative popularity of headsets in any given month, I could just read off the figures in the survey results. However, to compare the Quest 2 and Quest 3’s four-month sales to each other, I had to use player counts from SteamDB and make a few assumptions.
The first assumption is that the Steam Hardware Survey’s data is consistent for all users. Because Steam users have to opt-in to the survey, when it says that 2.24% of Steam users used a VR headset in January 2024, what it really means is that 2.24% of Steam Hardware Survey participants used a VR headset that month. There’s no reason to believe the survey’s sample isn’t representative of the whole of Steam’s user base, and this is an assumption that’s generally taken for granted when looking at Hardware Survey data. But if I’m going to break down where my numbers come from, I might as well do it thoroughly.
Secondly, I had to assume that Steam users only used one VR headset each month and that they didn’t share their headsets with other Steam users. These assumptions allow me to say that if the Meta Quest 3 was used for 14.05% of Steam VR sessions, then 14.05% of Steam users with a VR headset (which is 2.24% of Steam’s total users) owned a Quest 3 in January 2024. Not making these assumptions leads to an undercount and overcount, respectively, so they kinda cancel each other out. Also, without this assumption, I couldn’t continue beyond this step as I’d lack the data I need.
Who needs more than one VR headset anyway? (Image credit: Shutterstock / agencies)
Valve doesn’t publish Steam’s total user numbers, and the last time it published monthly active user data was in 2021 – and that was an average for the whole year rather than for each month. It also doesn’t say how many people take part in the Hardware Survey. All it does publish is how many people are using Steam right now. This information is gathered by SteamDB so that I and other people can see Steam’s Daily Active User (DAU) average for January 2021 and January 2024 (as well as other months, but I only care about these two).
My penultimate assumption was that the proportion of DAUs compared to the total number of Steam users in January 2021 is the same as the proportion of DAUs compared to the total number of Steam users in January 2024. The exact proportion of DAUs to the total doesn’t matter (it could be 1% or 100%). By assuming it stays consistent between these two months, I can take the DAU figures I have – 25,295,361 in January 2024 and 24,674,583 in January 2021 – multiply them by the percentage of Steam users with a Quest 3 and Quest 2 during these months, respectively – 0.31% and 0.37% – then finally compare the numbers to one another.
The result is that the number of Steam users with a Quest 3 in January 2024 is 87.05% of the number of Steam users with a Quest 2 in January 2021.
My final assumption was that Quest headset owners haven’t become more or less likely to connect their devices to a PC to play Steam VR. So if it's 87% as popular on Steam four months after their respective launches, the Quest 3 has sold 87% as well as the Quest 2 did after their first four months on sale.
Windows 11 could soon run updates without rebooting, if the rumor mill is right – and there’s already evidence this is the path Microsoft is taking in a preview build.
This comes from a regular source of Microsoft-related leaks, namely Zac Bowden of Windows Central, who first of all spotted that Windows 11 preview build 26058 (in the Canary and Dev channels) was recently updated with an interesting change.
Microsoft is pushing out updates to testers that do nothing and are merely “designed to test our servicing pipeline for Windows 11, version 24H2.” The the key part is we’re informed that those who have VBS (Virtualization Based Security) turned on “may not experience a restart upon installing the update.”
Running an update without requiring a reboot is known as “hot patching” and this method of delivery – which is obviously far more convenient for the user – could be realized in the next major update for Windows 11 later this year (24H2), Bowden asserts.
The leaker has tapped sources for further details, and observes that we’re talking about hot patching for the monthly cumulative updates for Windows 11 here. So the bigger upgrades (the likes of 24H2) wouldn’t be hot-patched in, as clearly there’s too much work going on under the hood for that to happen.
Indeed, not every cumulative update would be applied without a reboot, Bowden further explains. This is because hot patching uses a baseline update, one that can be patched on top of, but that baseline model needs to be refreshed every few months.
Add seasoning with all this info, naturally, but it looks like Microsoft is up to something here based on the testing going on, which specifically mentions 24H2, as well.
Analysis: How would this work exactly?
What does this mean for the future of Windows 11? Well, possibly nothing. After all, this is mostly chatter from the grapevine, and what’s apparently happening in early testing could simply be abandoned if it doesn’t work out.
However, hot patching is something that is already employed with Windows Server, and the Xbox console as well, so it makes sense that Microsoft would want to use the tech to benefit Windows 11 users. It’s certainly a very convenient touch, though as noted, not every cumulative update would be hot-patched.
Bowden believes the likely scenario would be quarterly cumulative updates that need a reboot, followed by hot patches in between. In other words, we’d get a reboot-laden update in January, say, followed by two hot-patched cumulative updates in February and March that could be completed quickly with no reboot needed. Then, April’s cumulative update would need a reboot, but May and June wouldn’t, and so on.
As mentioned, annual updates certainly wouldn’t be hot-patched, and neither would out-of-band security fixes for example (as the reboot-less updates rely on that baseline patch, and such a fix wouldn’t be based on that, of course).
This would be a pretty cool feature for Windows 11 users, because dropping the need to reboot – to be forced to restart in some cases – is obviously a major benefit. Is it enough to tempt upgrades from Windows 10? Well, maybe not, but it is another boon to add to the pile for those holding out on Microsoft’s older operating system. (Assuming they can upgrade to Windows 11 at all, of course, which is a stumbling block for some due to PC requirements like TPM).
ChatGPT maker OpenAI has now unveiled Sora, its artificial intelligence engine for converting text prompts into video. Think Dall-E (also developed by OpenAI), but for movies rather than static images.
It's still very early days for Sora, but the AI model is already generating a lot of buzz on social media, with multiple clips doing the rounds – clips that look as if they've been put together by a team of actors and filmmakers.
Here we'll explain everything you need to know about OpenAI Sora: what it's capable of, how it works, and when you might be able to use it yourself. The era of AI text-prompt filmmaking has now arrived.
OpenAI Sora release date and price
In February 2024, OpenAI Sora was made available to “red teamers” – that's people whose job it is to test the security and stability of a product. OpenAI has also now invited a select number of visual artists, designers, and movie makers to test out the video generation capabilities and provide feedback.
“We're sharing our research progress early to start working with and getting feedback from people outside of OpenAI and to give the public a sense of what AI capabilities are on the horizon,” says OpenAI.
In other words, the rest of us can't use it yet. For the time being there's no indication as to when Sora might become available to the wider public, or how much we'll have to pay to access it.
(Image credit: OpenAI)
We can make some rough guesses about timescale based on what happened with ChatGPT. Before that AI chatbot was released to the public in November 2022, it was preceded by a predecessor called InstructGPT earlier that year. Also, OpenAI's DevDay typically takes place annually in November.
It's certainly possible, then, that Sora could follow a similar pattern and launch to the public at a similar time in 2024. But this is currently just speculation and we'll update this page as soon as we get any clearer indication about a Sora release date.
As for price, we similarly don't have any hints of how much Sora might cost. As a guide, ChatGPT Plus – which offers access to the newest Large Language Models (LLMs) and Dall-E – currently costs $ 20 (about £16 / AU$ 30) per month.
But Sora also demands significantly more compute power than, for example, generating a single image with Dall-E, and the process also takes longer. So it still isn't clear exactly how well Sora, which is effectively a research paper, might convert into an affordable consumer product.
What is OpenAI Sora?
You may well be familiar with generative AI models – such as Google Gemini for text and Dall-E for images – which can produce new content based on vast amounts of training data. If you ask ChatGPT to write you a poem, for example, what you get back will be based on lots and lots of poems that the AI has already absorbed and analyzed.
OpenAI Sora is a similar idea, but for video clips. You give it a text prompt, like “woman walking down a city street at night” or “car driving through a forest” and you get back a video. As with AI image models, you can get very specific when it comes to saying what should be included in the clip and the style of the footage you want to see.
To get a better idea of how this works, check out some of the example videos posted by OpenAI CEO Sam Altman – not long after Sora was unveiled to the world, Altman responded to prompts put forward on social media, returning videos based on text like “a wizard wearing a pointed hat and a blue robe with white stars casting a spell that shoots lightning from his hand and holding an old tome in his other hand”.
How does OpenAI Sora work?
On a simplified level, the technology behind Sora is the same technology that lets you search for pictures of a dog or a cat on the web. Show an AI enough photos of a dog or cat, and it'll be able to spot the same patterns in new images; in the same way, if you train an AI on a million videos of a sunset or a waterfall, it'll be able to generate its own.
Of course there's a lot of complexity underneath that, and OpenAI has provided a deep dive into how its AI model works. It's trained on “internet-scale data” to know what realistic videos look like, first analyzing the clips to know what it's looking at, then learning how to produce its own versions when asked.
So, ask Sora to produce a clip of a fish tank, and it'll come back with an approximation based on all the fish tank videos it's seen. It makes use of what are known as visual patches, smaller building blocks that help the AI to understand what should go where and how different elements of a video should interact and progress, frame by frame.
Sora starts messier, then gets tidier (Image credit: OpenAI)
Sora is based on a diffusion model, where the AI starts with a 'noisy' response and then works towards a 'clean' output through a series of feedback loops and prediction calculations. You can see this in the frames above, where a video of a dog playing in the show turns from nonsensical blobs into something that actually looks realistic.
And like other generative AI models, Sora uses transformer technology (the last T in ChatGPT stands for Transformer). Transformers use a variety of sophisticated data analysis techniques to process heaps of data – they can understand the most important and least important parts of what's being analyzed, and figure out the surrounding context and relationships between these data chunks.
What we don't fully know is where OpenAI found its training data from – it hasn't said which video libraries have been used to power Sora, though we do know it has partnerships with content databases such as Shutterstock. In some cases, you can see the similarities between the training data and the output Sora is producing.
What can you do with OpenAI Sora?
At the moment, Sora is capable of producing HD videos of up to a minute, without any sound attached, from text prompts. If you want to see some examples of what's possible, we've put together a list of 11 mind-blowing Sora shorts for you to take a look at – including fluffy Pixar-style animated characters and astronauts with knitted helmets.
“Sora can generate videos up to a minute long while maintaining visual quality and adherence to the user’s prompt,” says OpenAI, but that's not all. It can also generate videos from still images, fill in missing frames in existing videos, and seamlessly stitch multiple videos together. It can create static images too, or produce endless loops from clips provided to it.
It can even produce simulations of video games such as Minecraft, again based on vast amounts of training data that teach it what a game like Minecraft should look like. We've already seen a demo where Sora is able to control a player in a Minecraft-style environment, while also accurately rendering the surrounding details.
OpenAI does acknowledge some of the limitations of Sora at the moment. The physics don't always make sense, with people disappearing or transforming or blending into other objects. Sora isn't mapping out a scene with individual actors and props, it's making an incredible number of calculations about where pixels should go from frame to frame.
In Sora videos people might move in ways that defy the laws of physics, or details – such as a bite being taken out of a cookie – might not be remembered from one frame to the next. OpenAI is aware of these issues and is working to fix them, and you can check out some of the examples on the OpenAI Sora website to see what we mean.
Despite those bugs, further down the line OpenAI is hoping that Sora could evolve to become a realistic simulator of physical and digital worlds. In the years to come, the Sora tech could be used to generate imaginary virtual worlds for us to explore, or enable us to fully explore real places that are replicated in AI.
How can you use OpenAI Sora?
At the moment, you can't get into Sora without an invite: it seems as though OpenAI is picking out individual creators and testers to help get its video-generated AI model ready for a full public release. How long this preview period is going to last, whether it's months or years, remains to be seen – but OpenAI has previously shown a willingness to move as fast as possible when it comes to its AI projects.
Based on the existing technologies that OpenAI has made public – Dall-E and ChatGPT – it seems likely that Sora will initially be available as a web app. Since its launch ChatGPT has got smarter and added new features, including custom bots, and it's likely that Sora will follow the same path when it launches in full.
Before that happens, OpenAI says it wants to put some safety guardrails in place: you're not going to be able to generate videos showing extreme violence, sexual content, hateful imagery, or celebrity likenesses. There are also plans to combat misinformation by including metadata in Sora videos that indicates they were generated by AI.
1 844.742.7448 
Email Us
Remote Support 
Contact
